It’s no secret – artificial intelligence is all the rage in the association industry. Every conference and trade show has at least a handful of sessions on the subject, and most any technology solution provider is working to incorporate some level of generative machine learning into their existing platforms, Those that haven’t done it already, at any rate. Industry publications are telling stories of how association pros are wanting to or succeeding in incorporating AI into their day-to-day. Which is great! For an industry that has been at times on the back foot in innovation, seeing this enthusiasm for change and groundbreaking tech is encouraging. It’s important to be realistic though. As incredible as AI evangelists make the tech seem, there’s the dream of what’s promised and there’s the reality of what’s possible, and associations need to reckon with that.

The promise of what artificial intelligence tools as they currently exist can do is certainly intriguing. Whether it’s more straightforward things like content creation – having ChatGPT write a blog post for you – or assistance with optimizing events and the member experience, there’s hope that associations can save time and deliver on the value proposition like never before. Not to mention all the “agents” that software tools like Salesforce are launching, chatbots that can answer questions clients or members have without having to pull a human from their mission-critical work. Salesforce promises that AgentForce can help with account updates, knowledge management (providing FAQs and educational materials to those who want it) or upselling, effectively being a valuable digital member of the staff. Pretty great!

According to Amex GBT Meetings & Events’ Global Meetings and Events Forecast and a recent article in Associations Now, a core AI for the organizations dabbling in it is matchmaking between sponsors and attendees. Keeping sponsors happy has always been of paramount importance for associations, and this makes that whole process easier. Now, attendees can be automatically linked up based on their various attributes, and sponsors can know ahead of time what to expect. Others are looking at sustainability improvements assisted with AI as well as discerning a quality location or venue. These insights and goals are communicated mainly through survey so just how this is being done is unclear, but it remains true that event organizers are bullish on what AI can do.

Similarly, leadership sees artificial intelligence as a great tool to bolster their organizational stewardship. A recent Salesforce study noted that executives believe much of the value in AI comes from efficiencies that can drive productivity growth. In essence, time saved, one of the constant needs of any association. As mentioned above, Some industry execs have the dream of creating some kind of automated chatbot with all their historical data in there, acting as a 24-hour helpline for members with questions. Much the way people ask ChatGPT and other LLMs questions about Superman versus Captain Marvel or to write a story involving Walter White and Tom Brady solving crimes Scooby-Doo style, this would be a tool that helps members solve questions and problems within their professional day-to-day.

Many AI tools have great potential value, but quite simply there’s no silver bullet out there. On top of that, typically, when people start considering incorporating these technologies into their operations, there can be little beyond “doing AI”, with no clear follow-up. There are tools out there with great promise, and companies like Open AI claim they’re going to save the world in some way or other, but the real returns aren’t exactly SkyNet.

The content creation aspect can be an attractive prospect, but it runs the risk of de-legitimizing whatever organization leans too heavily on it. Associations, being largely legacy organizations with a reputation built on decades of stewardship and service, need to protect that reputation. More and more we read articles about “AI slop” that is filling up the Internet, AI-produced blogs and listicles or procedurally generated images that are completely fake, overwhelming social media feeds and aggregators. An article completely written by a machine – even backed by some kind of real information – lacks the insight of something written by an expert or even just a layman who has put thought into it. Hallucinations, the things these chatbots make up, can be potentially problematic at the least and dangerous at the worst. Stories of chatbots offering false information on discounts, Google’s AI tool telling people to eat rocks and other variously strange things, these are real issues that could hurt an association that isn’t careful. Turning content creation wholly over to the machine could be seen as a great option in the near term, but not considering the long-ranging impact will be a detriment to those that go down that road.

From an executive standpoint, there’s the undermining of the staff that could be an issue, as well. Many workers see AI as a valuable tool, but at the same time see it as a potential job eliminator. Recent scientific studies have shown that the prospect of this has real and marked impacts on the well-being and effectiveness of current employees. So while executives might read an article or two about the supposed benefits, or want to make a leap to being more AI driven, the reality can be damaging to the internal morale of the organization. Adding new tools to save time, but losing considerable institutional knowledge, can

Another very real impact is, if you want to use one of these popular LLMs, you risk the valuable data that associations are supposed to protect. The old adage of “if it’s free, you’re the product” has only been codified as the 21st century has worn on, and tools like Claude.ai and ChatGPT are no different. Sure, they’ll charge you money if you want to pay, but there’s nothing stopping them from including whatever is imported into their training models. Whether it’s malicious actors or hackers stealing it, the data being used for other peoples’ answers (the New York Times is suing Open AI for just this case) or their simply ignoring whatever they promised to protect, it’s not always the best choice to simply trust them blindly.

Artificial intelligence is a fascinating next step in the digital evolution our lives are facing. There’s a tremendous opportunity here for associations to be on the front foot, and see new ways to bolster their efforts. As earlier noted, the time saving potential alone is attractive enough to spend at least some time exploring what’s out there. But there is no simple solution, nothing is magic, and there are still major questions looming for the technology that need answers. Future conferences will be rife with sessions focusing on what could be, and as the tools we all use start implementing their own versions of AI, associations will have their opportunities to explore. Doing so with a weathered eye and a dash of skepticism though, that’s the smart approach.